Dual-pulse excited linear prediction for speech coding

Abstract

The invention proposed a Dual-Pulse Excitation Model; wherein two pulses of each pair of pulses are always adjacent each other. Only one position index for each pair of pulses needs to be sent to the decoder, which saves bits to code all pulse positions. The magnitudes of each pair of pulses have limited number of patterns. Because the two pulses are adjacent each other, each pair of pulses with different magnitudes can produce different high-pass and/or low-pass effect. Since the magnitudes have enough variation, it is possible to assign the candidate positions of each pair of pulses within a small range in order to save the searching complexity.

1. A speech or signal coding method for encoding a signal, the coding method comprising:
 coding an excitation or a fixed codebook excitation wherein the excitation or the fixed codebook excitation includes plurality of pulse pairs called a Dual Pulse Model; 
 wherein said Dual Pulse Model features two pulses of each pair of pulses that are always adjacent to each other with a distance of 1, the two pulses of each pair of pulses have different magnitudes and signs and only one position index for each pair of pulses are transmitted from encoder to decoder. 
 
     
     
       2. The method of  claim 1 , wherein the Dual Pulse Model is used as a portion of popular CELP technology. 
     
     
       3. The method of  claim 1 , further comprising the steps of:
 selecting a best position of each pair of pulses within a limited set of candidate positions and only one best position index for each pair of pulses is sent to said decoder; 
 wherein possible magnitudes of each pair of pulses have enough variation so that the candidate positions of each pair of pulses can be limited in a relatively small range and a low complexity searching approach of the best pulse pair can be employed with local error minimization. 
 
     
     
       4. The method of  claim 3 , wherein the possible magnitudes of each pair of pulses are designed so that said pairs of pulses produce different high-pass effect. 
     
     
       5. The method of  claim 3 , wherein the possible magnitudes of each pair of pulses are designed so that said pairs of pulses produce different low-pass effect. 
     
     
       6. The method of  claim 3 , wherein the possible magnitudes of each pair of pulses are designed so that said pairs of pulses produce different low-pass and high-pass effect. 
     
     
       7. The method of  claim 3 , wherein the possible magnitudes of each pair of pulses are designed so that said pairs of pulses comprises the magnitude pattern of (1,−0.2), (0.5,−0.2), (1,−0.85) and (0.5,−0.85). 
     
     
       8. The method of  claim 3 , wherein the candidate positions of at least one pair of pulses cover the position candidate set {0, 1, 2, 3, 4, 5, 6, 7}. 
     
     
       9. The method of  claim 3 , wherein the best position and magnitude of each pair of uses can be jointly searched.